M. WIeser, W. aescHBacH-HertIg, t. scHneIder Institute of environmental physics,
university of Heidelberg, Heidelberg, germany
r.d. desHpande, s.K. gupta physical research laboratory, ahmedabad, India
Abstract
a quantitative palaeoclimate record was derived from the aquifer system of the cam-bay Basin, gujarat, India, a region characterised by a semi arid, monsoon dominated cli-mate. Stable isotopes generally increase with flow distance and 14c age, whereas noble gas temperatures (ngts) show a decline with age, amounting to a difference of ~3.5°c between Holocene and last glacial samples. the paper focuses on the Holocene covariation of the cli-mate proxies. Stable isotopes and excess air show consistent variations, confirming their interpretation as proxies for palaeohumidity. a group of early to mid Holocene samples de-pleted in stable isotopes and enriched in excess air indicates a phase of strong monsoon during the Holocene climate optimum. this is followed by a drying trend in the second half of the Holocene, and more humid conditions in the youngest part of the record. a temporary rise of NGT in the dry late Holocene may reflect a change in the soil temperature–air tem-perature relationship.
1. IntroductIon and study area
the monsoonal climates of asia, e.g. the Indian summer monsoon, affect the livelihoods of billions of people and are therefore a focus of climate research.
Many palaeoclimatic studies have detected a connection between northern hemi-sphere insolation and the intensity of the Indian summer monsoon [1], as the ItcZ is forced further towards the north with stronger insolation (e.g. ref. [2]). Implications of such relationships for the future may be a strengthening of the monsoon as a result of global warming.
our study region in the cambay Basin, northern gujarat, India, is semi arid and receives most of its precipitation from the Indian summer monsoon. Besides the mon-soon months, precipitation in gujarat originates only from isolated thunderstorm
Wieser et al.
events. Most rainfall takes place in the eastern highlands of the cambay Basin, feeding a sedimentary aquifer system in which flow is directed towards the southwest. The sed-iments filling the basin descend from both fluvial and aeolian origin, the youngest lay-ers being of Quaternary age. these young sediments have a thickness ranging from 300 to 800 m in their deeper parts [3]. shallow aquifers in most of the region have dried up due to exploitation of the groundwater over several decades, and deeper aquifers have a greater salt content due to rock–water interaction. the geographical distribution of the chemical and isotopic properties of groundwater is reported to be related to past climate change, the topography and the tectonic framework of the study area.
the palaeoclimate of India has been described by numerous studies. In late pleistocene times from 30 to 50 kyr Bp, dune and alluvial deposits [4] near the pro-ject location indicate an enhanced monsoon that declined around peak glacial times.
In the last glacial Maximum (lgM), little precipitation and arid conditions are in-dicated, as the alluvial deposits from flood plain aggradation make room for aeolian deposits. stalagmite data support these discoveries [5, 6]. after the end of the lgM, a strong monsoon phase occurred in the early Holocene and left marks in lake [7] and river [4] sediments and their isotopic condition. stalagmites in oman [2, 8] china [6]
and India [9] with stable isotope data also show enhanced precipitation, which is also predicted by model simulations [10]. In all these different studies, the climate opti-mum ranges between 10 to 12 and 5 to 8 kyr Bp.
temperature information is not provided by isotopic data in this subtropic re-gion, where the amount effect dominates stable isotope signals [2, 11]. noble gas temperatures (ngts) can provide quantitative temperature information, but only few such records are available for asia. the most adjacent studies of this type were con-ducted in oman, indicating a present–lgM temperature change of 6.5°c [12] and in china with a result of 4.5°c [13] also indicating an enhanced monsoon during the early Holocene.
the wet phase is superseded by a weakening of the monsoon which set in grad-ually [8]. desiccated lakes [7] and piston core data from the arabian sea as well as a change in vegetation and civilisation indicate this dry period with events of drought. a noble gas study from niger [14] also provided evidence for this dry phase in the late Holocene. several records place this time period from around 5 to 8 until 1 to 2 kyr BP. According to some records, the recent time shows more fluctuation [11]
with dry phases around 1 to 2 kyr Bp and a wet phase which started in about 600 year Bp. the most recent increase of monsoon precipitation may among other reasons re-sult from global warming [15].
2. MetHods
from 2008 to 2010, three sampling campaigns along two different tran-sects through the aquifer system of the north cambay Basin took place (fig. 1).
IAEA-CN-186/29
The transects running from the northeast towards the southwest follow the flow di-rection and slope of the basin, beginning with very shallow wells in the recharge area, where the aquifer is still unconfined and infiltration takes place, and ending with deep wells in the confined aquifer.
Parameters analysed directly in the field were water temperature, specific con-ductivity, alkalinity, dissolved oxygen and pH. stable isotopes and tritium were sam-pled in small glass bottles. the analysis of stable water isotopes was performed on a Mat 252 finnigan mass spectrometer and tritium was analysed radiometrically, both at the university of Heidelberg. 14c samples were taken in glass bottles and poisoned to avoid biological alteration. In the lab, dissolved carbon was extracted as co2 and converted to graphite, which was analysed by aMs at etH Zurich. for the noble gas samples, a copper tube fit into an aluminium rack was well flushed before being closed with steel clamps on either end. the samples were analysed on a gV 5400 noble gas mass spectrometer in Heidelberg. In this system, noble gases
FIG. 1. Map of the study area showing the positions of the sampled wells along two transects.
The 14C age of the water is indicated by the shading of the symbols. It increases progressively from the recharge area in the northeast towards the confined parts of the aquifer in the southwest. Thin lines indicate isolines of hydraulic head, bold lines outline tectonic faults, in par ticular the east and west Cambay Basin Bounding Faults (double lines) running NW–SE, perpendicular to groundwater flow.
Wieser et al.
are extracted from a sample, collected, cleaned and finally measured successively in the mass spectrometer. air aliquots are used for calibration.
gas solubility in water depends on the equilibration temperature. for inert gas-es, this can be used to identify the infiltration temperature of water masses from past times, as no temporal alteration takes place. noble gas concentrations extracted from water samples in a confined aquifer, therefore, reflect the water table temperature that prevailed in the recharge area at the time of infiltration. A complication occurs due to the trapping and (partial) dissolution of air bubbles during groundwater infiltration, leading to an excess of dissolved gas concentrations above solubility equilibrium.
This so-called excess air component is usually expressed by ΔNe, which is the neon excess in relation to the equilibrium concentration of neon, given in percent: ΔNe (%)
= (neexc/neeq) × 100%.
Various models to parameterise the excess air component have been pro-posed [16]. ΔNe is almost independent of the adopted model and depends mainly on the pressure exerted on the bubbles, which in turn is due to increases of the wa-ter table. In arid regions, several studies have attributed high excess air signals to a strongly fluctuating water table as a result of strong rain events between periods of aridity [14, 17]. the effect is more pronounced in regions with temporal aridity such as monsoon affected regions, but can also be observed in temperate zones [18].
therefore, excess air appears to be an interesting climate proxy for changes in the de-gree of precipitation.
the equilibration temperature (ngt), the amount of excess air and its possible fractionation relative to the composition of atmospheric air are three model param-eters, which can be determined by fitting the model to the observed concentrations of the four atmospheric noble gases ne, ar, Kr, Xe [19]. Helium is ignored in such calculations, as non-atmospheric components occur.
3. results and dIscussIon
groundwater was dated by 14c, taking into account various correction mod-els for the addition of old carbon from the sediments. The final ages were estimated as the mean of selected models, which yielded broadly consistent ages. the main characteristics of the age distributions are displayed in fig. 1. firstly and as expect-ed, the ages of the wells increase with increasing distance to the recharge area in the northeast. secondly, ages increase gradually on the northern transect, while on the southern transect the ages first increase slowly east of the west Cambay Basin Bounding fault, but increase rather quickly west of it. ages of all wells in the un-confined region east of the east Cambay Basin Bounding Fault are very young, as indicated by the presence of bomb-derived tritium.
Stable isotopes ratios δ18O and δ2H show a general increase with age. Higher values of δ18o during the lgM are at least in part attributed to the global ice-volume
IAEA-CN-186/29
effect. In contrast, the variability of δ18o of about 1‰ during the Holocene is prob-ably mostly due to the amount effect, indicating changes in monsoon strength [10].
this interpretation is supported by a comparable pattern of variation in the excess air (ΔNe) signal. The variation of δ18O and ΔNe for the Holocene samples from the con -fined part of the aquifer (the uncon-fined area is assumed to be too strongly disturbed by effects of point recharge and pumping) is shown in fig. 2. Based on the 14c ages and the characteristics of various tracers, three groups of Holocene samples were de-fined, which are distinguished in this plot: samples from the early Holocene climate optimum, from the following period of the late Holocene, and a few modern samples.
High excess air signals are found in the samples from the climate optimum and the modern period. These samples also show generally lower δ18o values. Both char-acteristics are indicative of humid conditions, i.e. phases of strong monsoon activity.
stronger precipitation and massive rain events in these times are therefore strongly suggested and agree with numerous palaeoclimate studies of the region. In contrast, the late Holocene group of intermediate age shows generally low excess air amounts and comparatively high δ18o values. Weak monsoon and dry conditions are there-fore expected in this period, again confirmed by other studies [7]. The independent humidity proxies provided by the stable isotopes and the noble gases thus confirm the view that a moist phase in the early Holocene was followed by dryer conditions except for the most recent part of the Holocene.
It is interesting to note that a correlation between NGT and ΔNe seems to exist for the Holocene samples (fig. 3). contrary to expectations, the warmest ngts are FIG. 2. Excess air (as ΔNe) versus δ18O for Holocene samples. High ΔNe and depleted δ18O both indicate humid conditions in the early Holocene climate optimum as well as in modern times. In contrast, the late Holocene period shows low ΔNe and enriched δ18O, indicating dryer conditions.
Wieser et al.
not found in the climate optimum or modern times, but during the dry late Holo-cene epoch. the ngts of the late HoloHolo-cene group (28.7°c on average) are somewhat higher than the present-day mean annual air temperature (Maat) of ahmedabad of 27.5°c. In contrast, the ngts obtained for the two more humid Holocene phases are in good agreement with modern Maat. We interpret the variation of ngt be-tween the Holocene groups as a change in soil temperature rather than air tempera-ture. The noble gas thermometer directly reflects the soil temperature at the water table, which is in general close to Maat. However, in dry periods such as the late Holocene in our study area, soil temperatures may be significantly higher than air temperatures, producing an offset in the relationship between ngt and Maat.
the same effect of enhanced soil and thus noble gas temperatures relative to air temperature in dry periods has been found before in a study in niger [14]. It was explained by reduced vegetation in dry periods exposing the soil to radiative heating.
That study also found a concurrent relationship between δ18O and ΔNe, identifying the dry and humid phases. We conclude from these results that in warm semi arid regions such as the sahel zone and north gujarat, India, ngt is a good proxy for air temperature only in comparatively wet periods. In dry phases, ngt is enhanced compared to Maat.
this result has to be taken into account in the estimation of temperature chang-es between the last glacial period and the Holocene from noble gas data. Based on δ18O and ΔNe data, we find that the LGM in our study area was a rather dry phase, consistent with other climate records. therefore, the ngt for this period should be FIG. 3. Excess air (as ΔNe) versus NGT for Holocene samples. Low ΔNe values indicating dry conditions in the late Holocene period coincide with high NGTs, presumably due to a relative enhancement of soil temperatures compared to air temperatures in dry phases.
IAEA-CN-186/29
compared to the warmest ngts from the dry late Holocene phase in order to avoid offsets due to changes in the air–soil temperature relationship. this comparison results in an estimate for the difference in mean annual air temperature between the Holocene and the last glacial period of 3.5 ± 0.5°c.
a slightly higher temperature change of 4.5°c has been observed in northeast-ern china [13]. as shown by palaeotemperature studies in different latitudes [16]
temperature change between Holocene and the lgM was more pronounced at higher latitudes. Our result for northwest India fits into this pattern. It is, however, contra-dictory to a groundwater study from oman, which suggested a tempera ture change between modern times and the lgM of 6.5°c [12] at a location relatively close to the present study site. It may be crucial for groundwater studies in tropical arid re-gions to take changes of the air–soil temperature relationship due to changing humid-ity into account.
REFERENCEs
[1] KutZBacH, J.e., Monsoon climate of the early Holocene: climate experiment with the earth’s orbital parameters for 9000 years ago, science 214 (1981) 59–61.
[2] fleItMann, d., Burns, s.J., MangInI, a., Mudelsee, M., KraMers, J., VIlla, I., neff, u., al-suBBary, a.a., Buettner, a., HIppler, d., Mat-ter, a., Holocene ItcZ and Indian monsoon dynamics recorded in stalagmites from oman and yemen (socotra), Quat. sci. rev. 26 (2007) 170–188.
[3] desHpande, r.d., groundwater in and around cambay Basin, gujarat: some geochemical and Isotopic Investigations, phd thesis, physical research laboratory, ahmedabad (2006).
[4] Juyal, n., cHaMyal, l.s., BHandarI, s., BHusHan, r., sIngHVI, a.K., continental record of the southwest monsoon during the last 130 ka: evidence from the southern margin of the thar desert, India. Quat. sci. rev. 25 (2006) 2632–
2650.
[5] fleItMann, d., Burns, s.J., neff, u., MangInI, a., Matter, a., chang-ing moisture sources over the last 330 000 years in Northern Oman from fluid-inclu-sion evidence in speleothems, Quatern. res. 60 (2003) 223–232.
[6] Wang, y., cHeng, H., edWards, l., Kong, X., sHao, X., cHen, s., Wu, J., JIang, X., Wang, X., an, Z., Millennial- and orbital-scale changes in the east asian monsoon over the past 224 000 years, nature 451 (2008) 1090–1093.
[7] enZel, y., ely, l.l., MIsHra, s., raMesH, r., aMIt, r., laZar, B., raJag-uru, s.n., BaKer, V.r., sandler, a., High-resolution Holocene environmen-tal changes in the thar desert, northwestern India, science 284 (1999) 125–128.
[8] fleItMann, d., Burns, s.J., Mudelsee, M., neff, u., KraMers, J., MangInI, a., Matter, a., Holocene forcing of the Indian Monsoon recorded in a stalagmite from southern oman, science 300 (2003) 1737–1739.
[9] raMesH, r., High resolution Holocene monsoon records from different proxies:
an assessment of their consistency, current science 81 (2001) 1432–1436.
Wieser et al.
[10] lee, J.-e., sWann, a.l., evaluation of the “amount effect” at speleothem sites in the asian monsoon region, earth environ. sci. 9 (2010).
[11] yadaVa, M.g., raMesH, r., Monsoon reconstruction from radiocarbon dated tropical Indian speleothems, Holocene 15 (2005) 48–59.
[12] WeyHenMeyer, c.e., Burns, s.J., WaBer, H.n., aescHBacH-HertIg, W., KIpfer, r, looslI, H.H., Matter, a., cool glacial temperatures and chang-es in moisture source recorded in oman groundwaters, science 287 (2000) 842–845.
[13] KreuZer, a. M., Von roHden, c., frIedrIcH, r., cHen, Z.y., sHI, J.s., HaJdas, I., KIpfer r., aescHBacH-HertIg, W., a record of temperature and monsoon intensity over the past 40 kyr from groundwater in the north china plain, chem. geol. 259 (2009) 168–180.
[14] Beyerle, u., rueedI, J., leuenBerger, M., aescHBacH-HertIg, W., peeters, f., KIpfer, r., dodo a., evidence for periods of wetter and cooler climate in the sahel between 6 and 40 kyr Bp derived from groundwater, geophys.
res. lett. 30 (2003) 1173.
[15] anderson, d.M., oVerpecK, J.t., gupta, a.K., Increase in the asian south-west Monsoon during the past four centuries, science 297 (2002) 596–599.
[16] KIpfer, r., aescHBacH-HertIg, W., peeters, f,. stute, M., “noble gas-es in lakgas-es and ground waters”, noble gasgas-es in geochemistry and cosmochemistry (porcellI, d., BallentIne, c., WIeler, r., eds), Mineralogical society of america, geochemical society. Washington, dc, rev. Mineral. geochem. 47 (2002) 615–700.
[17] aescHBacH-HertIg, W., Beyerle, u., HolocHer, J., peeters, f., KIp-fer, r., “excess air in groundwater as a potential indicator of past environmen-tal changes", study of environmenenvironmen-tal change using Isotope techniques (proc. Int.
conf. Vienna), 2001, Iaea, Vienna (2002).
[18] IngraM, r.g.s., HIscocK, K.M., dennIs, p.f., noble gas excess air applied to distinguish groundwater recharge conditions, environ. sci. technol. 41 (2007) 1949–1955.
[19] aescHBacH-HertIg, W., peeters, f., Beyerle, u., KIpfer, r., Interpre-tation of dissolved atmospheric noble gases in natural waters, Water resour. res. 35 (1999) 2779–2792.